The goals of this proposal are to determine 1) the degree of nonlinearity of human tissues, 2) how this information can be exploited to obtain reliable ultrasound images, and finally 3) what type of information nonlinearity can provide about the nature and disease of tissues. We have recently described a technique to measure the nonlinear parameter B/A by measuring the phase of sound transmitted through tissues. For better precision and reliability we will computerize the technique and use it to detemrine B/A for a large number of excised human tissues. Each sample will also be analyzed for fat-water composition and a correlation, if any, will be established between B/A and tissue composition. Also, ultrasound computer-assisted tomography will be utilized to obtain images of acoustic nonlinearity by measuring the perssure amplitude of the difference fequency generated from the propagation of two coincident primary waves of different frequencies. The results of this study will enable us to evaluate the potential role of nonlinear acoustics in biomedicine.